Extreme ultraviolet (“EUV”) light, e.g., electromagnetic radiation having wavelengths of around 5-100 nm or less (also sometimes referred to as soft x-rays), and including light at a wavelength of about 13 nm, can be used in photolithography processes to produce extremely small features in substrates, e.g., silicon wafers.
Methods to produce EUV light include, but are not necessarily limited to, converting a material into a plasma state that has an element, e.g., xenon, lithium or tin, with an emission line in the EUV range. In one such method, often termed laser produced plasma (“LPP”) the required plasma can be produced by irradiating a target material, for example in the form of a droplet, stream or cluster of material, with a laser beam.
For this process, the plasma is typically produced in a sealed vessel, e.g., vacuum chamber, and monitored using various types of metrology equipment. In addition to generating EUV radiation, these plasma processes also typically generate undesirable by-products in the plasma chamber which can include out-of-band radiation, high energy ions and debris, e.g., atoms and/or clumps/microdroplets of the target material.
These plasma formation by-products can potentially heat, damage or reduce the operational efficiency of the various plasma chamber optical elements including, but not limited to, collector mirrors including multi-layer mirrors (MLM's) capable of EUV reflection at normal incidence and/or grazing incidence, the surfaces of metrology detectors, windows used to image the plasma formation process, and the laser input window. The heat, high energy ions and/or debris may be damaging to the optical elements in a number of ways, including coating them with materials which reduce light transmission, penetrating into them and, e.g., damaging structural integrity and/or optical properties, e.g., the ability of a mirror to reflect light at such short wavelengths, corroding or eroding them and/or diffusing into them. Thus, it is typically desirable to minimize the amount of and/or the effect of plasma generated debris.
Heretofore, LPP systems have been disclosed in which each droplet is irradiated by a separate laser pulse to form a plasma from each droplet. Also, systems have been disclosed in which each droplet is sequentially illuminated by more than one light pulses. In some cases, each droplet may be exposed to a so-called “pre-pulse” and a so-called “main pulse”, however, it is to be appreciated that more than one pre-pulse may be used and more than one main pulse may be used and that the functions of the pre-pulse and main pulse may overlap to some extent. Typically, the pre-pulse(s) may function to expand the material and thereby increase the amount of material which interacts with the main pulse and the main-pulse may function to convert most or all of the material into a plasma and thereby produce an EUV light emission. However, it is to be appreciated that the functions of the pre-pulse and main pulse may overlap to some extent, e.g., the pre-pulse(s) may generate some plasma, etc. The increased material/pulse interaction may be due a larger cross-section of material exposed to the pulse, a greater penetration of the pulse into the material due to the material's decreased density, or both. Another benefit of pre-pulsing is that it may expand the target to the size of the focused pulse, allowing all of the pulse to participate. This may be especially beneficial if relatively small droplets are used as targets and the irradiating light cannot be focused to the size of the small droplet. Thus, in some applications, it may be desirable to use pre-pulsing to increase conversion efficiency and/or allow use of relatively small, e.g. so-called, mass limited targets. The use of relatively small targets, in turn, may be used to lower debris generation and/or reduce source material consumption.
As indicated above, one technique to produce EUV light involves irradiating a target material. In this regard, CO2 lasers, e.g., outputting light at infra-red wavelengths, e.g. 9.3 μm or 10.6 μm, may present certain advantages as a drive laser irradiating a target material in an LPP process. This may be especially true for certain target materials, e.g., materials containing tin. For example, one advantage may include the ability to produce a relatively high conversion efficiency between the drive laser input power and the output EUV power. Another advantage of CO2 drive lasers may include the ability of the relatively long wavelength light (for example, as compared to deep UV at 193 nm) to reflect from relatively rough surfaces such as a reflective optic that has been coated with tin debris. This property of 10.6 μm radiation may allow reflective mirrors to be employed near the plasma for, e.g., steering, focusing and/or adjusting the focal power of the drive laser beam.
Another factor worthy of consideration is the difficulty associated with consistently and accurately hitting a series of relatively small, quickly moving droplets with a pulsed laser beam at relatively high repetition rates. For example, some high-volume EUV light sources may call for the irradiation of droplets having a diameter of about 20-50 μm and moving at a velocity of about 50-100 m/s, at repetition rates exceeding 30 kHz.
In addition to the above described techniques, U.S. Pat. No. 6,855,943 (hereinafter the '943 patent) which issued to Shields on Feb. 15, 2005 and is entitled “DROPLET TARGET DELIVERY METHOD FOR HIGH PULSE-RATE LASER-PLASMA EXTREME ULTRAVIOLET LIGHT SOURCE” discloses a technique in which only some of the droplets in a droplet stream, e.g., every third droplet, is irradiated to produce a pulsed EUV light output. As disclosed in the '943 patent, the nonparticipating droplets (so-called buffer droplets) advantageously shield the next participating droplet from the effects of the plasma generated at the irradiation site. Unfortunately, in some cases, these buffer droplets may reflect light back into the laser causing self-lasing, which among other things, can reduce the effectiveness of the laser's gain media in producing high energy pulses. This may be especially true for high gain (e.g., G=1000−10,000) infra-red lasers, e.g., CO2 lasers, which tend to self-lase rather easily. Thus, it may be desirable to minimize any losses between pulses, including minimizing the effect of reflections. It may also be desirable to maximize gain extraction from a drive laser amplifier during pulse generation to produce large energy pulses, and in some cases, to provide stable consisting pulses, e.g. by minimizing instabilities, such as those caused by mode hopping in certain CO2 lasers having a MO-PA configuration.
With the above in mind, Applicants disclose a drive laser for an EUV light source.